********************************************************************************
* Margins Plots
********************************************************************************

clear
clear matrix
set more off
set scheme s1color
estimates clear
graph drop _all
set matsize 2500
log close _all
graph set window fontface "Times New Roman"

** Set Directory
cd "../Do"

********************************************************************************
** LOAD AND CLEAN DATA
********************************************************************************

use ../Data/matched_data_durables_jun2018_baseline.dta, clear

*labelling the 3 different durables spending variables; all are composites; one is nominal; another is "real" deflated using a single (appliances) cpi; another is "real" deflated on a good-by-good basis (in previous merge file)
la var durables "nominal durable goods spending"
la var durables_real1 "real durable goods spending, single deflator"
la var durables_real2 "real durable goods spending, separate deflators"

drop ethnicity

recode mort (5=0)
recode stocks (5=0)
recode retacct (5=0)
recode howner (5=0)

* Recode some expectations variables from the Inflation surveys. 

* unemployment dummies
recode q6 (1=1) (2 3 = 0), gen(unemp_increase)
recode q6 (3=1) (1 2 = 0), gen(unemp_decrease)

gen conditions_12m = q2a
* note everybody who says "other" is in separate category
recode q2a (1=1) (2 3 = 0), gen(conditions_12m_better)
recode q2a (2=1) (1 3 = 0), gen(conditions_12m_worse)

gen interestrate_12m = q7
recode interestrate_12m (1=1) (2 3 = 0), gen(intrate_12m_up)
recode interestrate_12m (3=1) (1 2 = 0), gen(intrate_12m_down)

gen bconditions_12m = q4
* note everybody who says "other" is in separate category
recode q4 (1=1) (2 3 = 0), gen(bconditions_12m_better)
recode q4 (2=1) (1 3 = 0), gen(bconditions_12m_worse)

* house price forecasts

gen hppoint = .
replace hppoint = 0 if q41==3
replace hppoint = q42 if q41==1
replace hppoint = -q42 if q41==2
* there are some extreme outliers (in the tens of thousands)
replace hppoint = . if abs(hppoint)>=200

la var hppoint "House price expectation"


*** Prepare some additional descriptives
* Race isn't reported in all periods. 
preserve
collapse (mean) race , by(prim_key)
sort prim_key
tempfile race
save `race'
restore

drop race
sort prim_key
merge m:1 prim_key using `race'
tab _merge
drop _merge
drop if prim_key==""

recode race (1=1) (nonmissing = 0), gen(white)
gen nonwhite = 1-white
recode gender (2=1) (1=0), gen(female)
recode highesteducation (4 9 = 0) (10/16 = 1), gen(coll)

* q31s* gives codes. Kind of odd that they've created separate variables. 
*MB: odd to generate employed variable, because everyone in sample is employed--however some also say they are retired (?) 
gen employed = . 
replace employed=1 if q31s1==1
replace employed = 0 if (q31s2==2 | q31s3==3 | q31s4==4 | q31s5==5 | q31s6==6 | q31s7==7) & q31s1!=1


* generate currently retired variable.
*odd that some are retired even though all are employed; how do retired people have wage growth expectations? they may be working a small job anyway
drop retired 
gen retired = . 
replace retired = 1 if q31s5==5
replace retired = 0 if (q31s1==1 | q31s2==2 | q31s3==3 | q31s4==4 | q31s6==6 | q31s7==7) & q31s5!=5
tab retired


*below also added December 2019 from Arman's old code
gen gas_expect = . 
replace gas_expect = 0 if q47a==3
replace gas_expect = q47a_higher if q47a==1
replace gas_expect = -q47a_lower if q47a==2

*more preparation for regressions
local expectations "intrate_12m_up intrate_12m_down unemp_increase unemp_decrease rw_expect d_wageiqr hppoint"
local infl1 "d_inflmedian d_infliqr"
local infl2 "d_longinflmedian d_longinfliqr"

*MBwhen using lagss--may need to rename lagged vars or these locals to match
*local lags1 "lagE_d_inflmedian lagE_d_infliqr" ;
*local lags2 "lagE_d_longinflmedian lagE_d_longinfliqr" ;

local spec1 "Short-Run Infl. Exp."
local spec2 "Medium-Run Infl. Exp."
la var d_inflmedian "Inflation Expectations"
la var d_infliqr "Inflation Uncertainty"

*MB resume: add "howner_fix" code ?

*replace howner=0 if howner==.
*adds ten people to homeowner status; there are still some with howner_fix==0 who have positive mortgage payment, but missing data for has mortgage and/or mortgage amount
*two observations have howner==0 and mort==1 (say they're not a homeowner but they say they have a mortgage); not important so not recoding them for now

drop if mort==0 & amtmort!=. & amtmort>100
*fixing the mort indicator to impute plausible values for people based on other information
replace mort = 0 if (howner!=1 | mortgage==0) & mort==.
replace mort = 1 if (howner==1 | (mortgage>0 & mortgage!=.)) & mort==.

*notice that amtmort may contradict mort dummy: one case such that mort=0 and amtmort if positive; one has mort==0 and amtmort==0; these are consistent but you might expect amtmort to be missing
*I would drop the first person alternatively with including them and using their stated mortgage amount; including for now
*there are 44 cases of those with mortgages for whom mortgage balance is missing--so I will just drop these observations from the sample

*the mortgage dummy is non-missing for over 2000 observations; however the mortgage balance (amtmort) is observed only for 990 people
*tab mort
*JIMIN: we could check results for everyone with non-missing mortgage dummy regardless of mortgage balance being observed; would have to drop mortgage balance from the regression

la var mort "Has Mortgage"

*MB: generating quasi-continuous income variable based on midpoint of ranges of annual household income variables (familyincome and familyincome_part2)
rename familyincome inc2
rename familyincome_part2 inc2_2
drop if inc2==.
*here is the income variable: "new_faminc"
gen new_faminc=.
replace new_faminc=2500 if inc2==1
replace new_faminc=6250 if inc2==2
replace new_faminc=8750 if inc2==3
replace new_faminc=11250 if inc2==4
replace new_faminc=13750 if inc2==5
replace new_faminc=17500 if inc2==6
replace new_faminc=22500 if inc2==7
replace new_faminc=27500 if inc2==8
replace new_faminc=32500 if inc2==9
replace new_faminc=37500 if inc2==10
replace new_faminc=45000 if inc2==11
replace new_faminc=55000 if inc2==12
replace new_faminc=67500 if inc2==13
replace new_faminc=87500 if inc2==14 & (inc2_2==1 | inc2_2==.)
*above accounts for one person with inc2==14 and inc2_2 missing; not sure why that's the case but I asigned them the lowest category of income over $75000
replace new_faminc=112500 if inc2==14 & inc2_2==2
replace new_faminc=162500 if inc2==14 & inc2_2==3
replace new_faminc=237500 if inc2==14 & inc2_2==4

gen log_new_faminc=log(new_faminc)

sort prim_key quarter
*bringing in SAMPLE WEIGHTS, to quarterly data: warning, we might lose observations if the set requiring weights has changed ; this will affect our ability to run regs using non-employed types (can only do unweighted) 
merge m:1 prim_key using ../Data/qweights_pooled
*these are the reg weights, the full sample weights will still be called weight_full
*weight variable is called "weight_samp" to indicate the weights were designed for the regression sample
*134 observations dropped that didn't merge with a weight_samp
drop if _merge!=3
drop _merge

*MB December 2019: key juncture: impose different sample restrictions and use different real/nominal durables
*********************
*define regression sample before recentering any variables: dropping extreme values
*below should drop top 2 highest values of durables spending; I doubt it makes any difference 
*JIMIN: try turning on and off the following restrictions
drop if durables>20000  | durables_real1>20000 | durables_real2>20000
drop if prim_key=="5041140:1"
drop if mortgage>200000 & mortgage!=.
drop if d_inflmedian>35
drop if d_longinflmedian>35
drop if hppoint<-50
**end of optional restrictions 


*assigning locals for weights (can turn on or off in regression)
local weights "[pweight=weight_samp]"
local weights_full "[pweight=weight_full]"
local pwfile "_pw"
*drop those with missing values for regressors
la var d_inflmedian "Inflation Expectation"
la var d_infliqr "Inflation Uncertainty"
la var d_longinflmedian "Inflation Expectation"
la var d_longinfliqr "Inflation Uncertainty"
la var lag_IE "Lagged Inflation Expectation"
la var lag_infl_iqr "Lagged Infl. Uncertainty"
la var hppoint "House price expectation"
la var gas_expect "Gas price expectation"

*variable for sum of monthly payments--interactions between this variable and IE may be included in some models
gen payments=mortgage+car if howner==1
replace payments=rent+car if howner!=1
gen log_payments=.
replace log_payments=log(payments) if payments>0
replace log_payments=0 if payments==0
*drop any observations with extreme value for payments (110,000): only if running a regression interacting with payments: actually none dropped here (unlike nondurables)
drop if payments>100000

*defining sample; removing those with missing values
drop if weight_samp==.
drop if d_inflmedian==.
drop if d_infliqr==.
drop if durables==.
*dropping lagged IE: only need if we include lag IE in regression 
drop if lag_IE==.
drop if lag_infl_iqr==.
*new income variable: new_faminc is recode of categorical variables familyincome and familyincome_part2; former variable was earnings last month and highly unreliable
drop if new_faminc==.
drop if intrate_12m_up==.
drop if intrate_12m_down==.
drop if unemp_increase==.
drop if unemp_decrease==.
drop if rw_expect==.
drop if d_wageiqr==.
drop if rage==.
drop if nonwhite==.
drop if female==.
drop if coll==.
drop if retired==.
drop if mort==.
*below results in loss of 300+ observations--results are robust not imposing this restriction and omitting hppoint from regressions
drop if hppoint==.
*below drops 111 observations--robustness applies again 
drop if howner==.


*replacing inflation expectations (and lags) with centered version: note that means are not weighted; this doesn't matter because weighted mean is within 0.04 ppts of unweighted mean IE
// egen IE_sampmean=mean(d_inflmedian) 
// replace d_inflmedian=d_inflmedian-IE_sampmean
egen lag_IE_sampmean=mean(lag_IE)
replace lag_IE=lag_IE-lag_IE_sampmean
egen lag_unc_sampmean=mean(lag_infl_iqr)
replace lag_infl_iqr=lag_infl_iqr-lag_unc_sampmean

*replacing log monthly payments with centered version
egen log_payments_sampmean=mean(log_payments)
replace log_payments=log_payments-log_payments_sampmean
*replacing log household income with centered version
egen log_inc_sampmean=mean(log_new_faminc)
replace log_new_faminc=log_new_faminc-log_inc_sampmean

sort prim_key quarter

*need to xtset the data
destring prim_key, generate(id_new) ignore(":")
xtset id_new quarter

*generate within-person means of time-varying independent variables: note that for variables that were recentered above, these now represent the within-household average of the deviation from the sample mean for the given variable
egen IE_bar=mean(d_inflmedian), by(id_new)
egen IE_unc_bar=mean(d_infliqr), by(id_new)
egen new_faminc_bar=mean(new_faminc), by(id_new)
egen log_new_faminc_bar=mean(log_new_faminc), by(id_new)
egen intrate_up_bar=mean(intrate_12m_up), by(id_new)
egen intrate_down_bar=mean(intrate_12m_down), by(id_new)
egen unemp_up_bar=mean(unemp_increase), by(id_new)
egen unemp_down_bar=mean(unemp_decrease), by(id_new)
egen rw_bar=mean(rw_expect), by(id_new)
egen rw_unc_bar=mean(d_wageiqr), by(id_new)
*egen hp_bar=mean(hppoint), by(id_new)
*egen howner_bar=mean(howner), by(id_new)
egen mort_bar=mean(mort), by(id_new)
egen log_payments_bar=mean(log_payments), by(id_new)
egen hp_bar=mean(hppoint), by(id_new)
egen howner_bar=mean(howner), by(id_new)
** Within-person mean of lagged IE and lagged infl uncertainty are not needed--they are collinear with within-person mean of current IE and uncertainty

*generate total durables spending within household
*add to below: use durables_real1 and durables_real2 instead--number of observations of "durables" per household will be identical to that for either "durables_real1" or "durables_real2", so no need to repeat below for alternate versions
egen tot_durables=total(durables), by(id_new)
sum tot_durables, d
*below drops 142 observations associated with households who never purchased durables under period of observation (reported zero spending on durables) 
drop if tot_durables==0
*generate variable that equals 1 in all cases, to sum to determine observations per person
gen pre_obs=1
*generate sum of observations per person
egen obs2=total(pre_obs), by(id_new)
sum obs2, d
*restrict on having nonzero durables spending in at least one period (based on total durables spending within household) 
*define sample  based on sufficient observation
gen durables_sample_hp=(obs2>=3)

** Reverse College
gen nocoll = 1-coll

gen boughtdur = (durcount >= 1 & !mi(durcount))

gen boughtdur2=(durables_real2>0 & durables_real2!=.)
la var durcount "Number of Durables Bought Including Cars"
la var boughtdur "Bought Durable including Car"
la var boughtdur2 "Bought Durable not including Cars"
count if boughtdur!=boughtdur2 & durables_sample_hp==1

********************************************************************************
** CREATE FIGURES
********************************************************************************

** Label variables 
la var log_new_faminc "Log Income"
la var log_new_faminc_bar "Mean Income (Log)"
la var IE_bar "Mean Inflation Exp. (SR)"
la var IE_unc_bar "Mean Inf. Uncertainty (SR)"
la var log_payments_bar "Mean Fixed Monthly Payments (Log)"
la var unemp_up_bar "Mean Unemp Increase"
la var unemp_down_bar "Mean Unemp Decrease"
la var intrate_up_bar "Mean Int Rates Increase" 
la var intrate_down_bar "Mean Int Rates Decrease"
la var rw_bar "Mean Real Wage Exp."
la var rw_unc_bar "Mean Wage Uncertainty"
la var hp_bar "Mean House Price Exp."
la var hppoint "House Price Exp."
la var howner_bar "Mean Homeowner"
la var mort_bar "Mean Has Mortgage"
la var mort "Has Mortgage"
la var nocoll "No College"
la var retired "Retired"
la var rage "Age" 
la var nonwhite "Non-White"
la var rw_expect "Real Wage Expectation"
la var d_wageiqr "Wage Uncertainty"
la var intrate_12m_up "Int Rates Inc"
la var intrate_12m_down "Int Rates Dec"
la var howner "Homeowner"
la var female "Female"
la var unemp_increase "Unemp Inc"
la var unemp_decrease "Unemp Dec"

local durablesv "Nom. Durable Spending"
local durables_real1v "Real Durable Spending, Single Deflator"
local durables_real2v "Real Durable Spending, Separate Deflator"

** FIGURE A3:

#delimit ; 
xtgee boughtdur c.d_inflmedian##i.nocoll c.d_infliqr##i.nocoll 
c.d_inflmedian##c.log_payments_bar c.d_inflmedian##c.mort_bar c.d_inflmedian##c.log_new_faminc_bar  log_payments log_new_faminc 
i.unemp_increase i.unemp_decrease i.intrate_12m_up i.intrate_12m_down 
rw_expect d_wageiqr i.mort i.nocoll rage i.nonwhite i.female i.retired i.howner hppoint
IE_bar IE_unc_bar log_new_faminc_bar intrate_up_bar intrate_down_bar 
unemp_up_bar unemp_down_bar rw_bar rw_unc_bar howner_bar hp_bar
i.quarter [pweight=weight_samp] if durables_sample_hp==1, 
family(binomial) link(logit) corr(exch) vce(robust);
#delimit cr

margins, eydx (d_inflmedian) over(nocoll) at(d_inflmedian=(1(1)7))

marginsplot, plotopts(connect(i)) title("") ytitle("") ///
	legend(order(3 "College" 4 "No College"))
graph export "../Figures/figureA3.png", as(png) replace 

set scheme s1mono

marginsplot, plotopts(connect(i)) title("") ytitle("") ///
	legend(order(3 "College" 4 "No College"))
graph export "../Figures/figureA3_bw.png", as(png) replace 


** FIGURE A4:

margins, eydx (d_inflmedian) at((p25) log_new_faminc (p25) log_new_faminc_bar) ///
	at((p50) log_new_faminc (p50) log_new_faminc_bar) ///
	at((p75) log_new_faminc (p75) log_new_faminc_bar) over(nocoll)
	
mat table = r(table)
	loc coll_25 = table[1,1]
	loc coll_ll_25 = table[5,1]
	loc coll_ul_25 = table[6,1]
	loc nocoll_25 = table[1,2]
	loc nocoll_ll_25 = table[5,2]
	loc nocoll_ul_25 = table[6,2]
	loc coll_50 = table[1,3]
	loc coll_ll_50 = table[5,3]
	loc coll_ul_50 = table[6,3]
	loc nocoll_50 = table[1,4]
	loc nocoll_ll_50 = table[5,4]
	loc nocoll_ul_50 = table[6,4]
	loc coll_75 = table[1,5]
	loc coll_ll_75 = table[5,5]
	loc coll_ul_75 = table[6,5]
	loc nocoll_75 = table[1,6]
	loc nocoll_ll_75 = table[5,6]
	loc nocoll_ul_75 = table[6,6]

clear
gen percentile = .
gen coll = .
gen coll_ll = . 
gen coll_ul = .
gen nocoll = . 
gen nocoll_ll = .
gen nocoll_ul = .

set obs 3
replace percentile = 25 in 1
replace percentile = 50 in 2
replace percentile = 75 in 3

foreach num in "25" "50" "75" {
	foreach var in "coll" "coll_ll" "coll_ul" "nocoll" "nocoll_ll" "nocoll_ul"  {
		replace `var' = ``var'_`num'' if percentile == `num'
	}
}	

set scheme s1color

tw (rcap coll_ll coll_ul percentile, lcolor(navy)) (rcap nocoll_ul nocoll_ll percentile, lcolor(cranberry)) ///
	(scatter coll nocoll percentile, color(navy cranberry)) , ///
	title("") xtitle("") ///
	ytitle("") xlab(25 "25th Percentile" 50 "50th Percentile" 75 "75th Percentile") xscale(range(10,90)) legend(order(3 "College" 4 "Non-College")) 
graph export "../Figures/figureA4.png", as(png) replace 

set scheme s1mono

tw (rcap coll_ll coll_ul percentile, lcolor(gray)) (rcap nocoll_ul nocoll_ll percentile, lcolor(black)) ///
	(scatter coll nocoll percentile, color(gray black)) , ///
	title("") xtitle("") ///
	ytitle("") xlab(25 "25th Percentile" 50 "50th Percentile" 75 "75th Percentile") xscale(range(10,90)) legend(order(3 "College" 4 "Non-College")) 
graph export "../Figures/figureA4_bw.png", as(png) replace 